OpenMMLab's Next Generation Video Understanding Toolbox and Benchmark

An open-source toolbox for action understanding based on PyTorch

[ECCV2024] Video Foundation Models & Data for Multimodal Understanding

Awesome papers & datasets specifically focused on long-term videos.

OpenTAD is an open-source temporal action detection (TAD) toolbox based on PyTorch.

Official Pytorch Implementation of 'Background Suppression Network for Weakly-supervised Temporal Action Localization' (AAAI-20 Spotlight)

[TIP 2022] End-to-end Temporal Action Detection with Transformer

A curated publication list on weakly-supervised temporal action localization

Official Pytorch Implementation of 'Weakly-supervised Temporal Action Localization by Uncertainty Modeling' (AAAI-21)

[ECCV 2022] Official Pytorch Implementation of the paper : " Zero-Shot Temporal Action Detection via Vision-Language Prompting "

[ICCV 2021] Relaxed Transformer Decoders for Direct Action Proposal Generation

Official Pytorch Implementation of 'Learning Action Completeness from Points for Weakly-supervised Temporal Action Localization' (ICCV-21 Oral)

[CVPR 2022] An Empirical Study of End-to-end Temporal Action Detection

[CVPR2021] CoLA: Weakly-Supervised Temporal Action Localization with Snippet Contrastive Learning

The official repository of our paper "ACM-Net: Action Context Modeling Network for Weakly-Supervised Temporal Action Localization".

[CVPR 2021] Multi-shot Temporal Event Localization: a Benchmark

TemporalMaxer: Maximize Temporal Context with only Max Pooling for Temporal Action Localization

Implementation for Bottom-Up Temporal Action Localization with Mutual Regularization (ECCV2020)

[CVPR 2022] Fine-grained Temporal Contrastive Learning for Weakly-supervised Temporal Action Localization

Adversarial Background-Aware Loss for Weakly-supervised Temporal Activity Localization (ECCV 2020)